Monitoring and Analysis of Wind Turbine Condition based on Multivariate Immunity Perception

Health monitoring of wind turbine (WT) has gained considerable attention, and the supervisory control and data acquisition (SCADA) data (such as power, temperature, and pressure, etc.) have been analyzed. However, many parameters in SCADA data are inertia parameters, which are easily affected by their inherent inertia. To solve this problem, a health monitoring method for WT based on multivariate immunity perception (MIP) is proposed in this paper. This method uses the immunity perception to obtain the health information of a single parameter and the analytic hierarchy process (AHP) to construct the immunity system. The weights of the immunity system are determined by fault statistics of a wind farm. The health condition of the subsystem/whole machine is quantified by combining the single parameter health information and immunity system. Two numerical examples are given to validate the method proposed in this paper.